Background The pipeline embolization device (PED) is approved for the treatment of large aneurysms of the proximal internal carotid artery (ICA). Its off-label application in treating aneurysms located specifically at the ICA terminus (ICA-T) has not been studied.
Methods We conducted a retrospective chart review of patients from 2011 to 7 treated with PEDs. Out of 365 patients, 10 patients with ICA-T aneurysms were included. Patient demographics, procedural information, follow-up imaging, and clinical assessments were recorded.
Results Mean age was 46.9 years (± 8.8), and 6 (60%) patients were women. The mean maximum diameter of the aneurysms treated was 14.7 mm (± 10.7) and the mean neck diameter was 9.3 mm (± 6.6). Reasons for presentation included six incidental findings, one acute subarachnoid hemorrhage (SAH), and three patients with prior SAH. Kamran–Byrne Occlusion Scale scores for the treated aneurysms were as follows: three class IV (complete obliteration), four class III (<50% filling in both height and width for fusiform aneurysms or residual neck for saccular aneurysms), one class II fusiform aneurysm, 1 class 0 saccular aneurysm (residual aneurysm body), and one not classified due to pipeline thrombosis. Two clinically asymptomatic complications were noted: one patient who had a small distal cortical SAH post PED and one patient whose stent was found to be thrombosed on follow-up angiogram. All patients were seen in follow-up, and no patients were found to have worsening of their pre-procedure modified Rankin Scale score.
Conclusion The PED has potential for treating ICA-T aneurysms not amenable to conventional treatment strategies. Further studies are warranted to confirm the long term outcomes.
- internal carotid artery terminus aneurysm
Statistics from Altmetric.com
The pipeline embolization device (PED) is a flow diversion stent used to effectively treat cerebral aneurysms of challenging anatomic subtypes.1–3 The first prospective multicenter trial confirming its comparable efficacy with traditional endovascular approaches and clipping was published in 2011 and it is currently approved by the Food and Drug Administration (FDA) for treating large and giant aneurysms of the cavernous and proximal supraclinoid internal carotid artery (ICA).1
While the off-label application of the PED in treating aneurysms at more distal segments of the cerebral vasculature is continuing to be evaluated, it usage specifically at the ICA terminus (ICA-T) has not yet been studied. Aneurysms at this location are known to be difficult to treat using traditional techniques due to the complexity of aneurysms that form at this bifurcation, especially in terms of large and perforator branch vessels.4
The study was approved by the local institutional review board. A retrospective review was conducted using a prospectively maintained database of patients who were treated with the PED from 2011 to 2017 at a single academic center. All patients were identified that received treatment for aneurysms at the ICA-T. Demographic information, aneurysm characteristics, treatment details, including antiplatelet regimen and follow-up data, were extracted from the medical record. Aneurysm occlusion was graded on the Roy–Raymond and Kamran–Byrne scales.5–8 All peri- and post-procedural complications were recorded.
All patients were premedicated with dual antiplatelet therapy. Platelet function tests were performed prior to all procedures as per the institution protocol. Optimal platelet reactivity was determined using light transmittance platelet aggregometry with a panel of four agonists (ADP 20 μM and 5 μM, arachidonic acid 500 μg/mL, and U46619 1 μM). Optimal platelet reactivity was determined using thresholds established by the cardiovascular literature for patients with acute coronary syndrome undergoing treatment with percutaneous coronary intervention. The thresholds used are high ADP 60, moderate ADP 40, and arachidonic acid 20. All patients demonstrated optimal platelet activity suppression prior to placement of a PED. Fellowship trained neuroendovascular surgeons performed all procedures. Intravenous heparin was administered intraprocedurally to achieve an activated clotting time >250 s. Heparin was discontinued on conclusion of the procedure. Dual antiplatelet therapy was continued for 6 months after PED placement.
The Shapiro–Wilk test was used to assess the normality of the variables. Continuous variables are reported as mean±SD or median (IQR), as appropriate. Categorical variables are reported as proportions. Statistical analysis was performed using IBM SPSS Statistics 24.
Mean age was 46.9 years (± 8.8) among a total of 10 patients (table 1). Six (60%) patients were women. The cohort was 50% White and 30% African American. The mean maximum diameter of the aneurysms treated was 14.7 mm (± 10.7) and the mean neck diameter was 9.3 mm (± 6.6). Reasons for presentation included six incidental findings, one acute subarachnoid hemorrhage (SAH) that was not amenable to coil embolization or clip ligation, and three patients with prior SAH who had recurrences of previously coiled aneurysms. All except for two patients received one PED with one patient requiring two PEDs and another patient requiring three PEDs. Five patients underwent coiling as well, including the patient who required two PEDs. All PEDs were positioned between the M1 and the supraclinoid segment of the ICA.
Follow-up imaging with DSA or MR angiography was conducted at an median of 7.5 months (IQR 14.3) post-PED placement. Kamran–Byrne Occlusion Scale scores for the treated aneurysms were as follows: three class IV (complete obliteration), four class III (<50% filling in both height and width for fusiform aneurysms or residual neck for saccular aneurysms), one class II fusiform aneurysm, and 1 class 0 saccular aneurysm (residual aneurysm body).8Raymond–Roy Occlusion Classifications for the treated aneurysms were as follows: three class I (complete obliteration), two class II (residual neck), and four class III (residual aneurysm body).5–7 One aneurysm was not classified on either scale due to stent thrombosis on follow-up angiogram. Aneurysm shape, typically classified as fusiform (figure 1) or saccular (figure 2), demonstrated a correlation to complete occlusion, with 3 (60%) saccular aneurysms versus no fusiform aneurysms having complete occlusion on follow-up imaging. Of the nine cases with DSA follow-up, three cases had angiographic non-filling of the anterior cerebral artery (ACA).
There were two complications noted. One patient presented with transient numbness and had a small distal cortical SAH post PED who remained at their neurologic baseline throughout. The patient was found to be a clopidogrel hyper-responder, with their high ADP decreasing from 64 to 23 and their moderate ADP decreasing from 46 to 16. A confirmatory PRU was also performed which showed 2 units. Accordingly, their clopidogrel dose was titrated to every other day dosing.
The other complication was a patient whose stent was found to be asymptomatically thrombosed from the distal supraclinoid ICA to the middle cerebral artery on follow-up angiogram. The patient had a known carotid occlusion on the contralateral side and asymptomatically filled their bilateral middle cerebral arteries from posterior temporal collaterals and was maintained on clopidogrel for prophylaxis. There was significant concern for thromboembolic complication, so the patient had confirmatory platelet testing 3 days post-procedure revealing an optimal high ADP response of 33 and an optimal moderate ADP response of 28.
All patients were seen in follow-up and no patients were found to have worsening of their pre-procedure modified Rankin Scale score. Despite PED placement completely covering the A1 origin in all cases, no strokes of the ACAwere noted on clinical examination or imaging.
This report describes the results of PED deployment for treating aneurysms located at the ICA-T at a single academic center. To our knowledge, this is the first study to examine clinical and radiographic outcomes for treated aneurysms specifically at this location. Bifurcations, such as the ICA-T, have been shown to have increased aneurysm formation due to unique flow dynamics.9 Aneurysms originating from the ICA-T itself are also often difficult to treat by conventional methods due to the anatomy. These aneurysms often require complex endovascular techniques, such as ‘Y stent’ assisted coiling or other endoluminal devices.10–13 A single center series of 37 patients with a meta-analysis of 158 patients demonstrated a retreatment rate of 25.7% for the single center series and 14% for the meta-analysis for ICA bifurcation aneurysms.14 Conversely, another study comparing surgical clipping with endovascular treatment, demonstrated only a 4% rate of major recanalization requiring treatment with a 42% rate of minor recanalization.15
For saccular aneurysms, flow diversion needs to be weighed against treatment with expanding myriad endovascular techniques. A large body of literature exists regarding coiling, and balloon and stent assisted coiling for bifurcation aneurysms, predominantly basilar apex, that is applicable to the ICA-T. Balloon assisted coiling is similarly safe relative to standard coiling with improved radiographic results.16 However, balloon remodeling is impotent against coil herniation in wide neck aneurysms and greatly diminishes tactile feedback during coil placement as the microcatheter is pinned by the balloon.
Bifurcation aneurysms are often amenable to stent assisted coiling. A large survey of papers covering 1517 patients found a 61% rate of occlusion on follow-up with a 19% overall complication rate and 9% rate of stent related technical complications. While these data likely overestimate actual risk with modern stents and experience, it does emphasize that new devices are initially associated with an increased rate of complication.17 Fargen et al reported a multicenter series of 45 patients with mostly basilar bifurcation aneurysms treated with ‘Y stenting’ with a Raymond–Roy I or II occlusion in 96% of cases and an 11% procedural complication rate.18 Building ‘Y stent’ constructs with ‘flow diversion-like properties’ has also been reported19
In terms of newer devices, Pulserider (Cerenovus, Irvine, California, USA) is a recently FDA approved device designed for coil assist embolization of bifurcation aneurysms. The ANSWER trial of bifurcation aneurysms (basilar n=27, ICA-terminus n=7) recently revealed Raymond–Roy I or II occlusion in 90% at 12 months with only one delayed ischemic event.13 Other devices for endovascular treatment of bifurcation aneurysms not FDA approved include pCONus (Phenox, Bochum, Germany) and Woven EndoBridge (WEB, Microvention, Aliso Viejo, California, USA). pCONus has shown promise, as an initial single center series showed 87% Raymond–Roy I or II occlusion with no permanent neurologic deficits or deaths.20
WEB aims to treat bifurcation aneurysms by intrasaccular flow diversion. A large study of 114 ruptured and unruptured aneurysms (10 ICA-T) treated with WEB at a single center demonstrated adequate aneurysm occlusion in 76% of cases at the last angiographic follow-up weighed against two cases of treatment associated morbidity and eight deaths. All deaths were related to antecedant SAH, not device placement.21 Another study compiled data from three large multicenter series, including 169 aneurysms (17 ICA-T) with 79% Raymond–Roy I or II occlusion at follow-up. Neurologic or device related morbidity was 3% at 1 month with no deaths, with two procedure related morbidities and one procedure related mortality at 1 year.22
Flow diversion uniquely modulates blood flow through the aneurysmal segment of a vessel without significantly interrupting flow to vital side branches1 and is particularly suited for treating wide neck and fusiform aneurysms for which traditional endovascular coiling is not an option. Flow diversion has not traditionally been used in the ICA-T due to apprehension over jailing the ACA origin. Nossek et al 23 demonstrated the safety of PED placement at the ICA-T in terms of ACA occlusion, and our study corroborates those data. Another study demonstrated slow flow in five PED covered ACAs on initial DSA and ACA narrowing on follow-up DSA. All patients were asymptomatic, likely due to a patent anterior communicating artery, which emphasizes the uniqueness of ACA coverage compared with other side branches.24 Importantly, the ipsilateral ACA should be at the largest codominant with the contralateral ACA when selecting ICA-T aneurysms to treat by flow diversion as the covered ACA–A1 is more likely to occlude, leading to aneurysm thrombosis.
Significant in our study is the lower occlusion rate for fusiform compared with saccular aneurysms. Overall, only 50% of aneurysms completely or near completely occluded, but 80% of saccular aneurysms achieved Raymond–Roy I or II occlusion. While confirmation by larger studies is needed, flow diversion is still an attractive option for fusiform aneurysms that lack other effective endovascular options and often require complex surgical solutions.
Lastly, many studies demonstrate the safety and efficacy of microsurgical clip ligation for saccular ICA-T aneurysms. Our center has the benefit of having multiple practitioners experienced in both modalities, and each of these aneurysms was considered for potential treatment via more traditional means. The decision to place a PED was multidisciplinary and made after considering a myriad of factors related to the patient’s anatomy and comorbidities. In general, we chose flow diversion as a firstline therapy in cases of fusiform aneurysm morphology, and only in select saccular cases. These cases were patients who were medically unfit for surgery or had multiple aneurysms treated with a single PED construct as well as cases where the potential for recurrence was quite high and would be lessened by placement of a flow diverter across a small or codominant ACA.
Our study has obvious limitations inherent to its retrospective design. While the results of this study are promising, the median follow-up for patients was only 7.5 months—we expect that aneurysm occlusion rates could increase with time. Additionally, the sample size of 10 patients was small, which was likely a result of the novelty of this treatment method, comparatively low occurrence of aneurysms at this location, and success of more traditional endovascular and surgical treatments of these aneurysms. Our continued experience, as well as studies at different institutions, will help to confirm the generalizability of these results.
The PED has potential for treating ICA-T aneurysms not amenable to conventional treatment strategies. These include most fusiform aneurysms as well as saccular aneurysms where the aneurysm morphology is thought to be conducive to flow diversion treatment. Further studies are warranted to confirm outcomes for long term follow-up.
AP and BMH contributed equally.
Contributors Conception and design: AP, BMH, and JAG. Acquisition of the data: AP, BMH, SLS, TPM, and JAG. Analysis and interpretation of the data: AP, BMH, and JAG. Drafting of the article: AP, BMH, and JAG. Critical revision: all authors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Ethics approval The study was approved by the Emory University institutional review board.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Any data analyzed for this study are available upon request from the corresponding author.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.